Metadynamics sampling in atomic environment space for collecting training data for machine learning potentials

作     者：Dongsun Yoo Jisu Jung Wonseok Jeong Seungwu Han 

作者机构：Department of Materials Science and Engineering and Research Institute of Advanced MaterialsSeoul National UniversitySeoulKorea Samsung Display Co.Ltd.1Samsung-roGiheung-guYongin-siGyeonggi-doKorea 

出 版 物：《npj Computational Materials》 (计算材料学（英文）)

年 卷 期：2021年第7卷第1期

页      面：1171-1179页

核心收录：

学科分类：12[管理学] 07[理学] 08[工学] 070203[理学-原子与分子物理] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 0805[工学-材料科学与工程（可授工学、理学学位）] 0835[工学-软件工程] 0701[理学-数学] 0811[工学-控制科学与工程] 0801[工学-力学（可授工学、理学学位）] 0702[理学-物理学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：This work was supported by Samsung Electronics(IO201214-08143-01) The computations were carried out at Korea Institute of Science and Technology Information(KISTI)National Supercomputing Center(KSC-2020-CRE-0125) 

主　　题：dynamics collective stretching 

摘      要：The universal mathematical form of machine-learning potentials(MLPs)shifts the core of development of interatomic potentials to collecting proper training ***,the training set should encompass diverse local atomic environments but conventional approaches are prone to sampling similar configurations repeatedly,mainly due to the Boltzmann *** such,practitioners handpick a large pool of distinct configurations manually,stretching the development period *** overcome this hurdle,methods are being proposed that automatically generate training ***,we suggest a sampling method optimized for gathering diverse yet relevant configurations *** is achieved by applying the metadynamics with the descriptor for the local atomic environment as a collective *** a result,the simulation is automatically steered toward unvisited local environment space such that each atom experiences diverse chemical environments without *** apply the proposed metadynamics sampling to H:Pt(111),GeTe,and Si *** these examples,a small number of metadynamics trajectories can provide reference structures necessary for training high-fidelity *** proposing a semiautomatic sampling method tuned for MLPs,the present work paves the way to wider applications of MLPs to many challenging applications.